How to use FileReader with datainputstream

[Bob Zoloman]

Hi, I am trying to use the FileReader with the DataInputStream to read in characters from a text file (ex .4RDE.2CS.9WD ). I've never actually worked with files before, so I'm completely clueless on how to do it, or if it can be done the way I am trying to do it.

I am using the '.' as a dilimiter, when the byte is a '.' I want to skip to the next byte and use that as a counter. The byte after the counter is the byte I want to repeat based on the counter value.

So for exampe .4R would be RRRR (4R's) and the rest would be DECCSWWWWWWWWWD

Some problems I am having are:
- What encoding should I save my test file as? ASCI, Unicode, UTF-8?
- If I try to use readChar() on the first character it returns a '?' character, but if I use readByte() it returns back a '.'? Any help is appreciated, thanks.

This is the code I am using:

  import java.io.*;
 
public class Decode
{
    
    public static void main(String[] args)
    {
        DataInputStream is = null;
        char ch;
        char delimiter = '.';
        short counter = 0;
        
        try
        {
            is = new DataInputStream(new FileInputStream(args[0]));
            
            while (true)
            { // exception deals catches EOF
                ch = is.readChar();
                
                if(ch == delimiter)
                {
                    counter = is.readShort();
                    //doesn't read in short value properly'
                }
                
                System.out.print((char) ch);
                System.out.flush();
            }
        }
        catch (EOFException eof)
        {
            System.out.println(" >> Normal program termination.");
        }
        catch (FileNotFoundException noFile)
        {
            System.err.println("File not found! " + noFile);
        }
        catch (IOException io)
        {
            System.err.println("I/O error occurred: " + io);
        }
        catch (Throwable anything)
        {
            System.err.println("Abnormal exception caught !: " + anything);
        }
        finally
        {
            if (is != null)
            {
                try
                {
                    is.close();
                }
                catch (IOException ignored)
                {
                }
            }
        }
    }
}
  

[Stan James]

I've never used DataInputStream but it looks as though it's only going to work with a file that was written by DataOutputStream or happens to have exactly the right binary format. readInt Doc You may have to use an inputstream, read one char at a time and parse the length char into an int.

Your compression scheme is very familiar. I've done it with a 2 or 3 character length field and gotten very good compression on fixed length COBOL fields padded with zeros & spaces. Be sure to handle a "." in your data!

[Bob Zoloman]

At the moment I'm using the FileInputStream to try and read it in one one char or short at a time, I'm not sure if that's the one I should be using with the DataInputStream. Anything that works is fine by me

The compression is a running length encoding.